Subcutaneous Insulin Safely Treats DKA, Keeps Patients From ICU

Miriam E. Tucker

April 07, 2022

A subcutaneous insulin-driven treatment protocol for diabetic ketoacidosis (DKA), instead of continuous intravenous insulin, can reduce intensive care unit (ICU) admission and hospital readmission in selected adult patients without an increase in adverse events, a new study finds.

Results from a 9-year real-world study involving nearly 8000 hospitalizations for DKA were published online April 7 in JAMA Network Open by Priya Rao, MD, hospitalist and chief of resource management at Kaiser Permanente San Jose Medical Center, California, and colleagues.  

"We found that implementing a subcutaneous insulin-driven protocol for treating diabetic ketoacidosis in the right patients was safe and was associated with significant changes in practice, including significantly decreased need for ICU and a reduction in readmission rate, with no increase in hypoglycemic events or mortality," Rao told Medscape Medical News.

Results were similar between the 30% of DKA patients with a diagnosis of type 2 diabetes and the 70% with type 1 diabetes, she noted.

Culture Change Was Biggest Challenge

In the United States, standard DKA therapy has been continuous intravenous regular insulin, with patients typically kept in the ICU based on the need for close monitoring by nursing staff. As a result, DKA treatment costs in the US exceed $5 billion annually.

Efforts have been made elsewhere to move DKA treatment out of the ICU, particularly in the United Kingdom, and most recently, during the COVID-19 pandemic.

There, national guidelines recommend a combination of fixed-rate IV infusion and subcutaneous insulin initially, followed by a transfer to subcutaneous insulin alone. Costs of DKA hospitalizations are reportedly lower.

"The ICU is a scarce resource that can become critically strained during seasonal influenza surge or as evidenced by the COVID-19 pandemic. Thus, many US critical care specialists have routinely identified DKA as a diagnosis well-suited for treatment in non-ICU settings if the right patients can be identified and appropriate treatment and monitoring can be put in place," Rao and colleagues write.

Culture change was the biggest challenge to establishing the subcutaneous DKA protocol, Rao said. "Our clinicians were used to treating DKA with IV insulin infusions and in the ICU...Our advice would be that other organizations establish a multidisciplinary team with physicians — including emergency department, hospitalists, intensivists, endocrinologists — nurses, and pharmacists when thinking through a subcutaneous insulin protocol for patients with DKA," she said.

Based on these study results, the subcutaneous protocol has since been adopted by the other 20 Northern California Kaiser Permanente Hospitals as of April 2021. The team will be performing a further analysis as well as exploring the use of continuous glucose monitoring to potentially facilitate the use of subcutaneous insulin for DKA management, Rao told Medscape Medical News.  

Protocol Involved Emergency, Non-ICU Ward, and Discharge

Starting in 2016, the intervention protocol, developed as a quality improvement project at Kaiser Permanente San Jose Medical Center, was established for nonpregnant adult patients with uncomplicated DKA.

Initial emergency department management includes administration of weight-based subcutaneous insulin glargine and lispro, along with IV lactated ringer boluses, dextrose 5% with normal saline infusion, and potassium repletion.

Management on the general medical-surgical wards focuses on ongoing volume expansion with lactated ringer fluid and weight-based lispro every 4 hours until the blood glucose level is at or below 250 mg/dL. The dextrose and saline are continued until patients are started on insulin dosing based on blood glucose measurements. Electrolytes are monitored and repleted as needed.

At discharge, an outpatient endocrinology consultation is requested for patients with new-onset type 1 diabetes.

For those who already have established diabetes providers, an electronic health record message is sent for post-discharge follow-up. Patients with type 2 diabetes are referred for chronic conditions program management.

Subcutaneous Protocol Improved Outcomes

Results from the protocol intervention period, January 1, 2017 to December 31, 2019, were compared with the pre-implementation period at the Kaiser San Jose institution from January 1, 2010 to December 31, 2015, and the results for both time periods were compared with data from 20 other Northern California Kaiser Permanente facilities, of which all relied on standard IV insulin protocols except one that adopted the subcutaneous protocol in January 2019.

A total of 7989 DKA hospitalizations occurred among 5046 patients, including 4739 hospitalizations (59.3%) that occurred pre-implementation of the protocol and 3250 (40.7%) that occurred after implementation. Patients were a mean age of 42.3 years, and 51.8% were female.

Prior to implementation, subcutaneous insulin was the first insulin administered in 40 intervention site hospitalizations (13.4%) versus 98 (80.3%) with the new protocol. In contrast, no significant change occurred at the standard care sites pre- versus post-implementation (14.7% vs 12.8%).

The proportion of hospitalizations directly admitted to ICU dropped at the protocol site from 67.8% pre-implementation to 27.9% post-implementation (P < .001). There was no difference in the proportion later admitted to ICU (P = .49).

In contrast, at the standard care sites, direct ICU admission actually increased over the same time period, from 75.6% to 79.5% of hospitalizations (P < .001).

After implementation, 30-day readmissions dropped from 21.1% to 9.8% (P = .006), whereas there was no change at the standard sites (17.8% vs 17.5% P = .75).

The proportions with hypoglycemia (serum glucose levels < 70 mg/dL at any time) dropped from 12.1% to 9.0% following protocol implementation at the intervention site and from 15.7% to 9.0% at the standard care sites.

After adjustments, reductions in both ICU admission and 30-day readmission at the intervention site compared with the standard sites were both significant, at 57% and 50%, respectively. The difference in length of stay, at 3%, was not significant.

The study was funded by The Permanente Medical Group, Kaiser Permanente Community Benefit, and the National Institutes of Health. The authors have reported no relevant financial relationships.

JAMA Netw Open. 2022;5:e226417. Full text

Miriam E. Tucker is a freelance journalist based in the Washington, DC, area. She is a regular contributor to Medscape, with other work appearing in The Washington Post, NPR's Shots blog, and Diabetes Forecast magazine. She is on Twitter: @MiriamETucker.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.